{"url":"/dataset/uspto-50k","name":"USPTO-50k","full_name":null,"description_markdown":"Subset and preprocessed version of Chemical reactions from US patents (1976-Sep2016) by Daniel Lowe.\r\nIt includes 50K randomly selected reactions that was later classified into 10 reaction classes by Nadine Schneider et al.","description_withheld":null,"homepage":"https://pubs.acs.org/doi/suppl/10.1021/acs.jcim.6b00564/suppl_file/ci6b00564_si_002.zip","introduced_date":"2016-11-22","introduced_date_note":null,"introduced_by":{"paper":"/paper/whats-what-the-nearly-definitive-guide-to","title":"What’s What: The (Nearly) Definitive Guide to Reaction Role Assignment","first_author":"Nadine Schneider","url":null},"license":null,"modalities":[],"tasks":[{"name":"Chemical Reaction Prediction","url":"/task/chemical-reaction-prediction","datasets_with_task":"/datasets/task/chemical-reaction-prediction"},{"name":"Single-step retrosynthesis","url":"/task/single-step-retrosynthesis","datasets_with_task":"/datasets/task/single-step-retrosynthesis"}],"languages":[{"name":"English","url":"/datasets/language/english"},{"name":"French","url":"/datasets/language/french"}],"variants":["USPTO-50k"],"data_loaders":[{"repo":"https://github.com/deepchem/deepchem","url":"https://github.com/deepchem/deepchem","frameworks":[]}],"num_papers_in_archive":60,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/single-step-retrosynthesis-on-uspto-50k","task":"Single-step retrosynthesis","dataset_variant":"USPTO-50k","rows":35,"metrics":["Top-1 accuracy","Top-3 accuracy","Top-5 accuracy","Top-10 accuracy","Top-20 accuracy","Top-50 accuracy"],"first_row_in_archive_order":{"model":"NAG2G (reaction class as prior)","paper":"/paper/node-aligned-graph-to-graph-generation-for","metrics":{"Top-1 accuracy":"67.2","Top-10 accuracy":"93.8","Top-3 accuracy":"86.4","Top-5 accuracy":"90.5"},"code_links":[{"title":"dptech-corp/nag2g","url":"https://github.com/dptech-corp/nag2g"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/ualign-pushing-the-limit-of-template-free","title":"UAlign: Pushing the Limit of Template-free Retrosynthesis Prediction with Unsupervised SMILES Alignment","date":"2024-03-25","rows_on_this_dataset":2,"code_links":1,"syntology":null},{"paper":"/paper/node-aligned-graph-to-graph-generation-for","title":"Node-Aligned Graph-to-Graph (NAG2G): Elevating Template-Free Deep Learning Approaches in Single-Step Retrosynthesis","date":"2023-09-27","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":3,"samples_ran":3,"samples_unverified":0,"pointer_only_for_licence":3,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/o-gnn-incorporating-ring-priors-into","title":"O-GNN: Incorporating Ring Priors into Molecular Modeling","date":"2023-05-01","rows_on_this_dataset":2,"code_links":1,"syntology":null},{"paper":"/paper/mars-a-motif-based-autoregressive-model-for","title":"MARS: A Motif-based Autoregressive Model for Retrosynthesis Prediction","date":"2022-09-27","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/g2gt-retrosynthesis-prediction-with-graph-to","title":"G2GT: Retrosynthesis Prediction with Graph to Graph Attention Neural Network and Self-Training","date":"2022-04-19","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/semiretro-semi-template-framework-boosts-deep-1","title":"SemiRetro: Semi-template framework boosts deep retrosynthesis prediction","date":"2022-02-12","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/chemformer-a-pre-trained-transformer-for","title":"Chemformer: a pre-trained transformer for computational chemistry","date":"2022-01-31","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/improving-few-and-zero-shot-reaction-template","title":"Improving Few- and Zero-Shot Reaction Template Prediction Using Modern Hopfield Networks","date":"2022-01-15","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/retrocomposer-discovering-novel-reactions-by","title":"RetroComposer: Composing Templates for Template-Based Retrosynthesis Prediction","date":"2021-12-20","rows_on_this_dataset":2,"code_links":1,"syntology":null},{"paper":"/paper/permutation-invariant-graph-to-sequence-model-1","title":"Permutation invariant graph-to-sequence model for template-free retrosynthesis and reaction prediction","date":"2021-10-19","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":6,"samples_ran":0,"samples_unverified":6,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/retroprime-a-diverse-plausible-and","title":"RetroPrime: A Diverse, plausible and Transformer-based method for Single-Step retrosynthesis predictions","date":"2021-09-15","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/deep-retrosynthetic-reaction-prediction-using","title":"Deep Retrosynthetic Reaction Prediction using Local Reactivity and Global Attention","date":"2021-08-05","rows_on_this_dataset":2,"code_links":1,"syntology":null},{"paper":"/paper/learning-graph-models-for-template-free-1","title":"Learning Graph Models for Template-Free Retrosynthesis","date":"2021-06-04","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/energy-based-view-of-retrosynthesis","title":"Energy-based View of Retrosynthesis","date":"2020-07-14","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/molecule-edit-graph-attention-network","title":"Molecule Edit Graph Attention Network: Modeling Chemical Reactions as Sequences of Graph Edits","date":"2020-06-27","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":6,"samples_ran":0,"samples_unverified":6,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/a-graph-to-graphs-framework-for","title":"A Graph to Graphs Framework for Retrosynthesis Prediction","date":"2020-03-28","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/augmented-transformer-achieves-97-and-85-for","title":"State-of-the-Art Augmented NLP Transformer models for direct and single-step retrosynthesis","date":"2020-03-05","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/retrosynthesis-prediction-with-conditional-1","title":"Retrosynthesis Prediction with Conditional Graph Logic Network","date":"2020-01-06","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":4,"samples_ran":0,"samples_unverified":4,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/molecular-graph-enhanced-transformer-for","title":"Molecular Graph Enhanced Transformer for Retrosynthesis Prediction","date":"2019-09-25","rows_on_this_dataset":3,"code_links":0,"syntology":null},{"paper":"/paper/computer-assisted-retrosynthesis-based-on","title":"Computer-Assisted Retrosynthesis Based on Molecular Similarity","date":"2017-11-16","rows_on_this_dataset":2,"code_links":1,"syntology":null},{"paper":"/paper/retrosynthetic-reaction-prediction-using","title":"Retrosynthetic reaction prediction using neural sequence-to-sequence models","date":"2017-06-06","rows_on_this_dataset":1,"code_links":1,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":4,"samples_harvested":19,"samples_ran":3,"samples_unverified":16,"pointer_only_for_licence":3,"papers_with_no_sample_that_ran":3,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}